Breast-feeding is recognised as the most appropriate method for feeding infants and is closely related to health during infancy and to chronic disease prevention in adulthood1, Reference Binns, Lee and Scott2. ‘Exclusive breast-feeding’ for the first six months of life and continued breast-feeding up to 2 years of age or beyond are recommended by the World Health Organization (WHO) and other authorities3.
There have been considerable changes in breast-feeding practices in China over the past 40 years. ‘Ever breast-fed’ rates in both urban and rural areas were over 80% in the 1950s and 1960s. But the rates started to decline considerably during the 1970s, especially in larger cities when the use of breast-milk substitutes became widespread. The ‘ever breast-fed’ rate in urban areas of China was 42.7% in 1975 and fell further to 33.6% in 1985Reference Wang, Zhu and Tong4–Reference Buxton, Gielen, Faden, Brown, Paige and Chwalow6. A survey from a rural area near Shanghai showed that the breast-feeding rate was 80% in the early 1980s and fell to 44% in the early 1990sReference Pang7. Thus the trends towards declining breast-feeding in urban and rural areas were similar; however, the rates in cities were lower.
To address the decline in breast-feeding, in the 1990s the Children's Development Plan of the Chinese government set a target to achieve a national ‘exclusive breast-feeding’ rate at 4 months of 80% by 2000Reference Niu, Zhao and Liu8. As a result of the many efforts introduced to promote breast-feeding, including the baby-friendly hospital initiative, women and child health protection legislation, social support programmes and breast-feeding education programmesReference Song9, the breast-feeding rate in China started to increase in the 1990sReference Zheng10. A survey undertaken in the Beijing Women's Hospital in 1990 showed that breast-feeding rates at 6 weeks were 79.2% in the rooming-in group and 39.8% in the control groupReference Huang, Cong and Zhang11. In 2002, a cross-sectional study undertaken in Beijing and four provinces (Shandong, Hubei, Zhejiang, Guangdong) showed that the ‘ever breast-fed’ rate was 90.1% and the ‘full breast-feeding’ rates at 4 and 6 months were 45.3% and 21.6%, respectively. The survey also documented that average duration of breast-feeding was 8.73 ± 4.21 months (mean ± standard deviation)Reference Zhang, Hao and Wang12. Another survey in five large cities (Guangzhou, Shanghai, Congqing, Xian and Changchun) in 2002 showed that the average breast-feeding duration was 8 months and the ‘any breast-feeding’ rates at 0, 4, 6, 12 and 24 months were 94.6, 61.0, 50.1, 5.3 and 0.4%, respectivelyReference Hu, Zhang, Li, Zhang and Feng13. From the report of the Third National Health Service Investigation in 2005, the breast-feeding initiation rates in urban and rural areas were 94.1 and 98.5%, respectively. The average breast-feeding duration in urban and rural areas was 5 and 6 months, respectivelyReference Bureau14. The duration of breast-feeding in this survey was shorter than in the previous two, but the reason for this was not explained.
These studies show that breast-feeding rates in China fell during the 1970s, to reach the lowest point in the 1980s and began to rise in the 1990s. The rates in urban areas were generally lower than in rural areas. However China is a huge country, geographically as well as in population size and ethnic diversity, and breast-feeding rates in different parts of China can vary considerablyReference Wang, Xu and Meng15, Reference Yun, Kang, Ling and Xin16.
The Xinjiang Uygur Autonomous Region (Xinjiang AR) is located in north-western China. Situated in the hinterland of the Eurasian continent, Xinjiang borders eight countries: Russia, Kazakhstan, Kyrgyzstan, Tajikistan, Pakistan, Mongolia, India and Afghanistan. By the end of 2004, the population of Xinjiang had reached 19.6 million. The region contains more than 13 ethnic groups, the Uygur accounting for 45.73%, the Han for 39.75% and the Kazakh for 7.04%. The birth rate was 16 per thousand and death rate 5.1 per thousand.
A cross-sectional study undertaken in Shihezi, Xinjiang, in 1994–1996 showed that the breast-feeding rate was 34% at 4 monthsReference Xiao17, well below Chinese and international targets. This had significant implications for the health of the children of the region. A more detailed study of breast-feeding was needed to provide the data necessary to implement a comprehensive health promotion programme. The aim of the present longitudinal study was to identify the prevalence of breast-feeding up until 6 months of age.
Methodology
A longitudinal cohort study of infant feeding practices was undertaken in Xinjiang AR, People's Republic of China. In this region, all babies are usually delivered in hospitals. Mothers who delivered babies during 2003 and 2004 were contacted while in hospital and were invited to participate in the study. After discharge, mothers were contacted in person or by telephone at approximately monthly intervals (at 0.5, 1.5, 2.5, 3.5, 4.5 and 6 months) using a structured questionnaire to obtain details of infant feeding practices.
A total of 1256 mothers were randomly recruited from five hospitals or institutes. In Shihezi People's Hospital, mothers were recruited every second day due to staffing constraints. In the other hospitals (Shihezi Maternal and Child Health Care Institute and Urumqi Maternal and Child Health Care Institute in urban areas; Chabuchaer Maternal and Child Health Care Institute and Yumin County Hospital in rural areas) all mothers were contacted. Almost all mothers who were contacted (1219 or 97%) agreed to participate. Urumqi is the capital city of Xinjiang where the Uygur ethnic group is in the majority, while Shihezi is a predominantly Han ethnic area. Chabuchaer and Yumin counties have a larger concentration of Kazakh people and other minorities.
The majority of participants, including the minority groups, could read and speak Chinese (Mandarin). The questionnaire was originally prepared in Mandarin, and was also translated into the Uygur language, which can also be understood by Kazakh mothers. For those who could not read Chinese, trained nurses who were fluent in the ethnic languages were available to help them complete the questionnaires. For all minority mothers, follow-up calls and visits were made in their own ethnic language by nurses from their own ethnic group.
The questionnaire was based on those developed by Scott, Binns and Duong that have been used extensively in breast-feeding cohort studies in Australia, Vietnam and KenyaReference Scott, Landers, Hughes and Binns18–Reference Lakati, Binns and Stevenson23. The questionnaires were designed to identify the feeding method and to collect information on factors associated with breast-feeding. After translation the questionnaires were tested in focus groups to ensure cultural appropriateness.
The project was approved by Xinjiang local research authorities (Shihezi University, Urumqi Science Research Committee) and the Human Research Ethics Committee of Curtin University, Australia. Mothers who agreed to participate in the study signed the consent page at the front of the questionnaire and were informed of their rights to withdraw from the follow-up process at any time without prejudice. All of the personal data collected were kept confidential.
All data analyses were carried out using the Statistical Package for the Social Sciences, release 12.0 (SPSS Inc.). Descriptive statistics and cross-tabulations were generated for demographic factors, life tables were used for breast-feeding rates and the Kaplan–Meier model was used to calculated mean duration of ‘exclusive breast-feeding’, ‘full breast-feeding’ and breast-feeding.
The definitions of breast-feeding used in this paper are as followsReference Binns24–Reference Binns and Scott26.
● Any breast-feeding: The infant receives breast milk (direct from the breast or expressed) with or without other drink, formula or other infant food.
● Exclusive breast-feeding: Breast-feeding while giving no other food or liquid, not even water, with the exception of drops or syrups consisting of vitamins, mineral supplements or medicine.
● Almost exclusive breast-feeding: The infant may receive small amounts of culturally valued supplements – water, water-based drinks, fruit juice and ritualistic fluids.
● Full breast-feeding: Includes ‘exclusive breast-feeding’ and ‘almost exclusive breast-feeding’.
Results
The details of the sample and the prevalence of the major demographic variables are shown in Table 1. The sample as recruited included 47% Han, 30% Uygur, 16% Kazakh and a small number of other minorities. Almost all of the mothers (1118) in the study were married, eight were separated and one was widowed.
Table 1 Demographic factors and breast-feeding percentage before discharge in Xinjiang, People's Republic of China, 2003–2004 (n=1219)
The breast-feeding rates were calculated by life table analysis and are detailed in Table 2. The breast-feeding initiation rate was 92% and by 6 months this had declined to 73%. The median duration of ‘exclusive breast-feeding’ was less than 1 month, of ‘full breast-feeding’ was 2.8 months, and of ‘any breast-feeding’ was greater than 6 months. As follow-up in this study was terminated at 6 months, the maximum duration of breast-feeding could not be determined. The average duration of ‘exclusive breast-feeding’ was 1.8 months (95% confidence interval (CI) 1.7–2.0) and of ‘full breast-feeding’ was 2.8 months (95% CI 2.7–2.9). At 6 months of age, 73% (CI 70.1–75.8) of infants were still receiving some breast milk.
Table 2 Breast-feeding rates (%) in Xinjiang, People's Republic of China, 2003–2004 (n=1219)
* 0 month refers to the time of discharge from hospital, which was generally one week postpartum.
Inevitably in a cohort study, some mothers were lost to follow-up. The percentage of mothers lost to follow-up at 0.5, 1.5, 2.5, 3.5, 4.5 and 6 months was 6, 7, 9, 12, 13 and 21%, respectively.
Mother's perceptions about breast-feeding duration and formula are shown in Table 3. The majority of mothers thought that breast milk was not enough for their baby's nutritional requirements by 6 months of age. Only 33% of mothers thought that breast milk could satisfy a baby's nutritional requirements at 6 months and 31% mothers planned to give their baby supplemental food before 4 months.
Table 3 Mother's perceptions about breast-feeding and formula in Xinjiang, People's Republic of China, 2003–2004 (n=1219)
The perceptions of the different ethnic groups were compared using binary logistic and multinomial logistic regression analysis, and more minority mothers were found to have incorrect perceptions about ‘exclusive breast-feeding’ duration and formula. For example, compared with Han mothers, minority mothers were less likely to consider that breast milk could satisfy a baby's nutritional requirements for more than 4 months (odds ratio (OR) = 0.4, 95% CI 0.2–0.5) and were more likely to plan giving their babies supplemental food before 4 months (OR = 14.6, 95% CI 10.2–21.1). Kazakh mothers more likely thought formula was a healthy food (OR = 3.1, 95% CI 1.3–8.1).
A comparison was made of the understanding of infant feeding of mothers from urban and rural areas. More mothers in rural areas (92.7%) thought that breast-feeding could satisfy the baby's requirements for more than 4 months compared with mothers in urban areas (80.4%) (χ2 = 22.7, P < 0.01). Fewer mothers in rural areas (6.8%) planned to start giving supplemental food to their babies before 4 months than mothers in urban areas (41.3%) (χ2 = 90.0, P < 0.01). Overall, 8.5% of mothers who were breast-feeding their babies were recommended to begin using additional infant food by medical staff and a similar percentage by other people, i.e. 17% in total.
Discussion
In Xinjiang AR, 92% of mothers were breast-feeding on discharge from hospital and 73% were still breast-feeding at 6 months. This is a considerable increase in the breast-feeding rate compared with the previous study in Shihezi in 1996, when the rate at 4 months was only 34%Reference Xiao17. The breast-feeding rate in Xinjiang at discharge was similar to that currently reported from Beijing and some inland provinces, and for breast-feeding at 6 months was higher than in those regionsReference Zhang, Hao and Wang12. The only Chinese province with a higher reported breast-feeding rate over the first six months was Tibet, but the ‘exclusive breast-feeding’ rate in Xinjiang was a little higher than that in TibetReference Dang, Yan, Wang, Zheng and Xie27. The breast-feeding initiation rate was a similar to that in AustraliaReference Scott, Landers, Hughes and Binns18, Reference Scott, Landers, Hughes and Binns28, which is higher than in Canada and the USAReference Dubois and Girard29, Reference Philipp, Merewood, Miller, Chawla, Murphy-Smith and Gomes30.
However, the ‘exclusive breast-feeding’ rate in Xinjiang was considerably lower than the target of 80% at 4 months set by the Chinese government. Many babies were given water, formula or cow's milk too early in their life. The early introduction of complementary foods may shorten breast-feeding durationReference Riva, Banderali, Agostoni, Silano, Radaelli and Giovannini31.
The main reason for the early introduction of water may be an inappropriate understanding of infant feeding by both medical staff and mothers. Feeding water was common in hospitals because many doctors thought that infant jaundice was associated with dehydration. Also, Xinjiang is a province where people live on the edge of desert and traditionally mothers feed baby water to their infants early in life. The early introduction of water is also common in TibetReference Dang, Yan, Wang, Zheng and Xie27. Because the majority of babies there were fed water, the ‘exclusive breast-feeding’ rate was very small and hence in that study more emphasis was placed on the ‘full breast-feeding’ rate in the analysis. The ‘full breast-feeding’ rate at 4–6 months was only 3.0% in urban areas and 12.8% in rural areas in TibetReference Dang, Yan, Wang, Zheng and Xie27. The ‘full breast-feeding’ rate in Shihezi was slightly higher than this at 7.9%.
The advertising and ready availability of infant formula could be factors in its early introduction to some infants. In this study, most mothers knew about formula and about one mother in 10 was recommended to use infant foods by the medical staff while she was still in hospital. Only one-third of mothers thought that breast-feeding could satisfy a baby's nutritional requirements until 6 months of age and there would be no need for other supplemental foods. Some Uygur mothers said: ‘Breast milk is not thick enough for baby's growth’. The majority of babies (65.1%) were fed formula as a complement to breast-feeding at 6 months of age. The average time for the introduction of solid foods in the Han and minority groups were 4.7 and 5.7 months, respectivelyReference Yu and Song32.
The main problem of breast-feeding in Xinjiang was the early introduction of formula or water. The WHO recommends that infants should be fed exclusively on breast milk from birth to 6 months of age, with no water or other liquids or food during this period33. Education should highlight ‘exclusive breast-feeding’ and correct some old misconceptions.
There are several limitations that need to be considered when interpreting the results of this study. While the sample included urban and rural areas, logistically it was not possible to include subjects from the very remote or nomadic areas of the province. However, only a small proportion of the population lives in these areas.
Conclusion
The present study documents infant feeding patterns in Xinjiang and highlights the main ways in which they differ from national and international best practices. ‘Any breast-feeding’ rates at discharge and at 6 months were 92.2 and 73.0%, respectively. ‘Exclusive breast-feeding’ rates at discharge and 6 months were 66.2 and 6.2%, respectively. The main problem of breast-feeding in Xinjiang was the early introduction of formula or water. Further studies are needed to identify factors associated with ‘exclusive breast-feeding’ and duration.
Acknowledgements
We gratefully acknowledge the willing assistance given by the mothers in our study, the hospital staff and nursing students. Without this assistance the study would not have been possible.
